A MRS (magnetic resonance spectroscopy or nuclear magnetic resonance NMR)-based PTSD (post-traumatic stress disorder) and mTBI (mild traumatic brain injury) diagnostic system and method uses MRS signals, already pre-processed by the MRS scanner software. The signals are collected in vivo from specific regions of the brain. A wavelet decomposition is applied to the MRS signals, and the amplitude of the wavelet coefficients and their location in the MRS signals are used as features for training diagnostic classifiers of disease states. These classifiers are identified through analysis of features of individuals whose health status is known. Once the classifiers are trained, patients can be diagnosed by using the same wavelet features extracted from in vivo MRS scans of their brain regions.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. A system as claimed in claim 1, wherein the training phase of the diagnostic tool is performed by analyzing MRS signals of subjects with PTSD and mTBI.
The system relates to a diagnostic tool for analyzing magnetic resonance spectroscopy (MRS) signals to detect and differentiate between post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI). The tool includes a training phase where MRS signals from subjects diagnosed with PTSD and mTBI are analyzed to establish a reference database. This database captures distinct spectral patterns associated with each condition, enabling the tool to identify biomarkers or signal characteristics that distinguish PTSD from mTBI. During the diagnostic phase, the tool processes MRS signals from a new subject and compares them against the trained database to determine the likelihood of PTSD, mTBI, or both. The system may also include preprocessing steps to enhance signal quality, such as noise reduction or normalization, and machine learning algorithms to improve classification accuracy. The tool aims to provide a non-invasive, objective method for diagnosing these conditions, addressing challenges in clinical differentiation where symptoms often overlap. The training phase ensures the system is calibrated to recognize subtle but consistent differences in MRS signals between PTSD and mTBI, improving diagnostic reliability.
3. A system as claimed in claim 1, wherein the diagnostic tool implements binary classifiers for PTSD and mTBI.
The system is designed for medical diagnostics, specifically for identifying post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI). These conditions often present overlapping symptoms, making accurate diagnosis challenging. The system uses binary classifiers to distinguish between PTSD and mTBI, improving diagnostic precision. The binary classifiers are trained on data that differentiates the physiological, psychological, or behavioral markers associated with each condition. The system may integrate data from multiple sources, such as patient-reported symptoms, neuroimaging, or physiological measurements, to enhance accuracy. By applying machine learning or statistical models, the classifiers generate probabilistic outputs indicating the likelihood of PTSD or mTBI. The system may also include user interfaces for inputting patient data and displaying diagnostic results, aiding healthcare providers in decision-making. The binary classification approach ensures clear, actionable insights, reducing misdiagnosis and improving patient outcomes. The system may be part of a larger diagnostic platform that includes additional tools for monitoring and treatment planning.
4. A system as claimed in claim 1, wherein the computer system trains diagnostic classifiers distinguishing healthy control subjects from those with PTSD and/or mTBI are trained using the subset of the wavelet features identified during the training phase.
The system pertains to medical diagnostics, specifically the detection of post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI) using machine learning. The core challenge addressed is the accurate and automated differentiation between healthy individuals and those affected by PTSD or mTBI, which is difficult due to overlapping symptoms and subjective diagnostic criteria. The system employs wavelet-based feature extraction to analyze physiological or neurological data, such as brain signals or other biomarkers, converting raw data into a set of wavelet features. During training, these features are used to develop diagnostic classifiers that distinguish between healthy control subjects and those with PTSD or mTBI. The training phase identifies the most relevant subset of wavelet features, which are then used to train the classifiers. This subset ensures that only the most discriminative features are retained, improving diagnostic accuracy and efficiency. The system leverages these trained classifiers to process new data, providing automated and objective diagnostic support. By focusing on wavelet features, the system enhances the reliability of PTSD and mTBI detection, addressing limitations in traditional diagnostic methods. The approach is particularly valuable in clinical settings where rapid and accurate diagnosis is critical.
5. A system as claimed in claim 1, wherein the diagnostic tool implements classifiers distinguishing healthy controls subjects from those with PTSD and/or mTBI learned in the training.
This system relates to a diagnostic tool for distinguishing between healthy individuals and those with post-traumatic stress disorder (PTSD) and/or mild traumatic brain injury (mTBI). The tool uses classifiers trained on data from control subjects and individuals with these conditions to identify patterns indicative of PTSD or mTBI. The classifiers are developed through a training process that involves analyzing physiological, behavioral, or cognitive data from a diverse set of subjects. The trained classifiers are then applied to new data to assess whether an individual exhibits symptoms consistent with PTSD, mTBI, or both. The system may incorporate multiple classifiers to improve accuracy and reliability, with each classifier potentially focusing on different aspects of the data. The diagnostic tool may be integrated into a broader healthcare system, allowing for automated or semi-automated assessment of individuals at risk. The goal is to provide an objective, data-driven method for early detection and diagnosis, reducing reliance on subjective assessments and improving treatment outcomes. The system may also include features for tracking changes in an individual's condition over time, enabling personalized monitoring and intervention strategies.
6. A system as claimed in claim 1, further comprising characterizing metabolites including N-acetylaspartate (NAA), creatine (Cre), choline (Cho), glutamate (Glu), glutamine (Gln), gamma-amino butyric acid (GABA), myo-inositol (mI), and lactate.
This invention relates to a system for analyzing brain metabolites using magnetic resonance spectroscopy (MRS). The system is designed to address the challenge of accurately quantifying key metabolic compounds in the brain, which is critical for diagnosing and monitoring neurological disorders. The system includes a magnetic resonance imaging (MRI) scanner configured to acquire spectroscopic data from a target region of the brain. The system further includes a processing module that processes the acquired data to characterize specific metabolites, including N-acetylaspartate (NAA), creatine (Cre), choline (Cho), glutamate (Glu), glutamine (Gln), gamma-amino butyric acid (GABA), myo-inositol (mI), and lactate. These metabolites are biomarkers for various neurological conditions, such as neurodegenerative diseases, psychiatric disorders, and brain tumors. The system may also include a display module to visualize the quantified metabolite levels, aiding clinicians in diagnosis and treatment planning. The processing module may employ advanced spectral analysis techniques, such as linear combination modeling or quantum estimation, to enhance the accuracy and reliability of metabolite quantification. The system may be integrated with existing MRI systems or operate as a standalone diagnostic tool.
7. A system as claimed in claim 1, wherein the subset of wavelet features with a highest performance for discrimination are determined by measuring performance as an average Percent Correct Classification (PCC) from multiple iterations of a k-fold cross-validation test of the wavelet features.
The system relates to signal processing and machine learning, specifically for feature selection in classification tasks using wavelet transforms. The problem addressed is the need to identify the most discriminative subset of wavelet features for accurate classification, ensuring optimal performance while reducing computational complexity. The system processes input signals by decomposing them into wavelet features through a wavelet transform. These features are then evaluated for their discriminative power in classifying the signals. The key innovation involves selecting the subset of wavelet features that exhibit the highest performance for discrimination. Performance is quantified by measuring the average Percent Correct Classification (PCC) across multiple iterations of a k-fold cross-validation test. This statistical approach ensures robustness by validating the features' effectiveness through repeated testing and partitioning of the data. The selected features are then used to train a classifier, improving accuracy and efficiency in signal classification tasks. This method is particularly useful in applications where signal variability and noise require robust feature selection, such as biomedical signal analysis, fault detection, or pattern recognition.
8. A system as claimed in claim 1, wherein the diagnostic tool employs the wavelet features at 3.87, 1.61 and 1.64 ppm to diagnose between both mTBI and PTSD and mTBI-only.
The system is designed for medical diagnostics, specifically distinguishing between mild traumatic brain injury (mTBI) and post-traumatic stress disorder (PTSD) in patients, as well as identifying mTBI-only cases. The system uses a diagnostic tool that analyzes wavelet features extracted from magnetic resonance spectroscopy (MRS) data. The key innovation lies in the specific chemical shift values at 3.87, 1.61, and 1.64 parts per million (ppm), which serve as biomarkers to differentiate between mTBI, PTSD, and mTBI-only conditions. The wavelet features are derived from the MRS signals, which provide quantitative measurements of metabolic activity in brain tissue. By processing these features, the system can classify patients into the correct diagnostic category with improved accuracy compared to traditional methods. This approach leverages advanced signal processing techniques to enhance diagnostic precision, addressing the challenge of overlapping symptoms between mTBI and PTSD, which often complicate clinical assessments. The system integrates these biomarkers into a diagnostic algorithm, enabling more reliable and objective evaluations for better patient management.
9. A system as claimed in claim 1, wherein the diagnostic tool employs the wavelet features at 1.29 ppm to diagnose between both mTBI and PTSD and PTSD-only.
The system is designed for medical diagnostics, specifically distinguishing between mild traumatic brain injury (mTBI) and post-traumatic stress disorder (PTSD), as well as identifying PTSD-only cases. The system uses a diagnostic tool that analyzes wavelet features extracted from medical data, particularly focusing on the 1.29 ppm spectral region. This region is significant for differentiating between mTBI and PTSD, as well as isolating PTSD-only cases. The wavelet features are derived from signal processing techniques applied to biological or medical data, such as magnetic resonance spectroscopy (MRS) or other imaging modalities. The diagnostic tool processes these features to generate a classification output, enabling clinicians to accurately diagnose the presence of mTBI, PTSD, or PTSD-only. The system enhances diagnostic precision by leveraging specific spectral markers, reducing misdiagnosis and improving patient outcomes. The method involves extracting wavelet features, analyzing the 1.29 ppm region, and applying classification algorithms to distinguish between the conditions. This approach addresses the challenge of overlapping symptoms between mTBI and PTSD, providing a more reliable diagnostic framework. The system integrates signal processing, machine learning, and medical imaging to deliver an advanced diagnostic solution.
10. A system as claimed in claim 1, wherein the training phase of the diagnostic tool is performed by analyzing MRS signals of subjects with PTSD and mTBI and wherein the diagnostic tool implements binary classifiers for PTSD and mTBI.
The system is designed for diagnosing post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI) using magnetic resonance spectroscopy (MRS) signals. The technology addresses the challenge of accurately distinguishing between these conditions, which often present overlapping symptoms, making diagnosis difficult with conventional methods. During the training phase, the diagnostic tool analyzes MRS signals from subjects known to have PTSD or mTBI to establish a reference dataset. The system then implements binary classifiers specifically trained to differentiate between PTSD and mTBI based on the spectral patterns in the MRS data. These classifiers are optimized to identify unique biochemical markers associated with each condition, improving diagnostic accuracy. The system may also include preprocessing steps to enhance signal quality and feature extraction techniques to highlight relevant spectral features. By leveraging MRS, which provides insights into metabolic changes in the brain, the system offers a non-invasive and objective diagnostic approach. The binary classification ensures clear differentiation between PTSD and mTBI, aiding clinicians in selecting appropriate treatment pathways. This method reduces reliance on subjective assessments and improves early intervention outcomes.
12. A method as claimed in claim 11, further comprising implementing binary classifiers for PTSD and mTBI.
The invention relates to a method for diagnosing and classifying psychological and neurological conditions, specifically post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI). The method involves analyzing physiological and behavioral data to detect and differentiate between these conditions. The core process includes collecting data from individuals, such as brain activity, cognitive performance, and emotional responses, and processing this data using machine learning techniques to identify patterns indicative of PTSD or mTBI. The method further incorporates binary classifiers specifically trained to distinguish between PTSD and mTBI, improving diagnostic accuracy. These classifiers use features derived from the collected data to generate probabilistic or deterministic outputs, enabling clinicians to make informed diagnostic decisions. The approach aims to address the challenge of overlapping symptoms between PTSD and mTBI, which often leads to misdiagnosis or delayed treatment. By leveraging advanced data analysis and machine learning, the method provides a more objective and efficient diagnostic tool for these conditions.
13. A method as claimed in claim 11, wherein diagnostic classifiers distinguishing healthy control subjects from those with PTSD and/or mTBI are trained using the final subset of the wavelet features.
The invention relates to a method for diagnosing post-traumatic stress disorder (PTSD) and/or mild traumatic brain injury (mTBI) using wavelet-based feature analysis. The method addresses the challenge of accurately distinguishing between healthy individuals and those with PTSD and/or mTBI by leveraging machine learning classifiers trained on wavelet-transformed physiological or neurological data. The method involves preprocessing raw data, such as electroencephalogram (EEG) signals, to extract time-frequency features using wavelet transforms. These features are then filtered to remove irrelevant or redundant data, resulting in a refined subset of wavelet features. The final subset is used to train diagnostic classifiers, which are optimized to differentiate between healthy control subjects and individuals with PTSD and/or mTBI. The classifiers may employ algorithms such as support vector machines, neural networks, or decision trees to achieve high diagnostic accuracy. By focusing on wavelet features, the method enhances the sensitivity and specificity of PTSD and mTBI detection, providing a non-invasive and objective diagnostic tool. The approach improves upon traditional methods by reducing noise and irrelevant data, leading to more reliable classification results. This technique is particularly valuable in clinical settings where early and accurate diagnosis is critical for effective treatment planning.
14. A method as claimed in claim 11, wherein the wavelet features with a highest performance for discrimination are determined by measuring performance as an average Percent Correct Classification (PCC) from multiple iterations of a k-fold cross-validation test of the wavelet features.
The invention relates to a method for selecting wavelet features in a classification system, particularly for improving discrimination performance in pattern recognition tasks. The method addresses the challenge of identifying the most effective wavelet features from a set of extracted features to enhance classification accuracy. Wavelet features are often used in signal processing and machine learning to represent data in a multi-resolution manner, but selecting the most discriminative features can be computationally intensive and time-consuming. The method involves evaluating wavelet features based on their classification performance. Specifically, the performance of each wavelet feature is measured by calculating an average Percent Correct Classification (PCC) from multiple iterations of a k-fold cross-validation test. K-fold cross-validation is a statistical technique used to assess how well a model generalizes to an independent dataset by partitioning the data into k subsets, training on k-1 subsets, and testing on the remaining subset. This process is repeated multiple times, and the PCC is averaged across all iterations to determine the most effective features for discrimination. The features with the highest average PCC are selected as the most discriminative wavelet features for the classification task. This approach ensures that the chosen features are robust and generalize well to unseen data, improving the overall accuracy of the classification system.
15. A method as claimed in claim 11, wherein the wavelet features at 3.87, 1.61 and 1.64 ppm are employed to diagnose between both mTBI and PTSD and mTBI-only.
This invention relates to a diagnostic method for distinguishing between mild traumatic brain injury (mTBI) and post-traumatic stress disorder (PTSD) using magnetic resonance spectroscopy (MRS). The method leverages wavelet features extracted from MRS data to identify specific chemical shifts in the brain. The key innovation involves analyzing wavelet features at three distinct proton resonance frequencies: 3.87 parts per million (ppm), 1.61 ppm, and 1.64 ppm. These frequencies correspond to metabolic markers that differ between individuals with mTBI and PTSD and those with mTBI alone. The method processes MRS data to extract these wavelet features, which are then used to classify the condition. This approach improves diagnostic accuracy by providing a quantitative, non-invasive method to differentiate between these overlapping conditions, addressing the challenge of misdiagnosis in clinical settings. The technique may be integrated into existing MRS systems for enhanced diagnostic workflows.
16. A method as claimed in claim 11, wherein the wavelet features at 1.29 ppm are employed to diagnose between both mTBI and PTSD and PTSD-only.
This invention relates to a medical diagnostic method using wavelet features extracted from magnetic resonance spectroscopy (MRS) data to distinguish between mild traumatic brain injury (mTBI), post-traumatic stress disorder (PTSD), and PTSD-only cases. The method leverages wavelet analysis to process MRS signals, specifically focusing on the spectral peak at 1.29 parts per million (ppm), which is associated with certain biochemical markers in the brain. By analyzing these wavelet features, the method enables differentiation between individuals with mTBI, those with PTSD, and those with PTSD but no history of brain injury. The technique improves diagnostic accuracy by providing a quantitative, signal-based approach to distinguish overlapping symptoms of mTBI and PTSD, addressing challenges in clinical assessment where these conditions often present similarly. The wavelet features are derived from MRS data, which is a non-invasive imaging technique that measures metabolic changes in brain tissue. This method enhances diagnostic precision, aiding in personalized treatment planning for patients with neurological and psychological trauma.
17. A method as claimed in claim 11, further comprising analyzing MRS signals of subjects with PTSD and mTBI and implementing binary classifiers for PTSD and mTBI.
This invention relates to the analysis of magnetic resonance spectroscopy (MRS) signals to diagnose and differentiate between post-traumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI). The method involves acquiring MRS signals from subjects suspected of having either condition. These signals are processed to extract metabolic biomarkers that correlate with PTSD or mTBI. The processed signals are then input into binary classifiers, which are trained to distinguish between the two conditions based on the extracted biomarkers. The classifiers output a diagnosis indicating whether the subject has PTSD, mTBI, or neither. The method may also include preprocessing steps to enhance signal quality, such as noise reduction and normalization. The classifiers are trained using labeled MRS data from known cases of PTSD and mTBI, ensuring accurate differentiation between the two conditions. This approach provides a non-invasive diagnostic tool to improve early detection and treatment planning for individuals with PTSD or mTBI.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 25, 2018
December 20, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.